Hae Yeon Kang, Kyungdo Han, Dae Hyun Yoon, Su-Yeon Choi, Seung Ho Choi
Marked BW change was associated with subsequent dementia risk in young and middle-aged adults with T2DM and depression. BW trajectories may serve as clinically observable risk markers, although associations involving extreme BW change should be interpreted cautiously.
AIM: Type 2 diabetes mellitus (T2DM) and depression are established risk factors for dementia, yet clinically observable markers of neurocognitive vulnerability in younger adults with these comorbidities remain poorly characterized. We examined whether longitudinal body weight (BW) change is associated with incident dementia in young and middle-aged adults with T2DM and depression.
MATERIALS AND METHODS: We analysed a nationwide cohort of 17 851 adults aged 20-59 years with T2DM and depression from the Korean National Health Insurance Service database. BW change between two health screenings (2010 and 2012) was categorized as < -10%, -10% to < -5%, -5% to +5% (reference), > +5% to < +10% and ≥ +10%. Participants were followed for a median of 10.3 years. Hazard ratios (HRs) were estimated using Cox proportional hazards models, with sensitivity analyses addressing competing risk of death, reverse causation, age dependency, depression burden and non-linearity.
RESULTS: During 178 342.2 person-years, 527 participants developed dementia (2.96 per 1000 person-years). Compared with stable BW, fully adjusted HRs were 1.53 (95% CI, 1.01-2.32) for < -10% BW loss, 1.69 (1.35-2.11) for -10% to < -5% loss, 1.41 (1.07-1.86) for > +5% to < +10% gain and 1.74 (1.13-2.68) for ≥ +10% gain. Restricted cubic spline analysis indicated a non-linear association (p for non-linearity = 0.0035). Findings were broadly consistent in analyses restricted to dementia before age 65, competing-risk analysis and participants aged 40-59 years. However, extreme BW loss and gain were attenuated in the 3-year lag analysis, and substantial BW loss was attenuated after adjustment for depression-related claim frequency.
CONCLUSIONS: Marked BW change was associated with subsequent dementia risk in young and middle-aged adults with T2DM and depression. BW trajectories may serve as clinically observable risk markers, although associations involving extreme BW change should be interpreted cautiously.